Background: Fragile X mental retardation syndrome (FXS), is the most common cause of mental retardation and the most common known cause of autism. Other symptoms of FXS include depression, sensory processing deficits, aggressive behavior, connective tissue problems and digestive difficulties. This disorder arises when the number of CGG-repeats in the 5 UTR of the FMR1 gene exceeds 200. Such alleles become silenced. This results in a deficiency of the protein product of this gene, FMRP, which is involved in the regulation of translation of a subset of mRNAs. The FMRP deficiency in brain results in aberrant dendritic spine morphology and defects in the response to synaptic activation. The mechanism of gene silencing is unknown.[unreadable] [unreadable] Progress report: We have taken 2 broad approaches. One has been to understand the factors important for normal FMR1 expression (Kumari and Usdin, 2001, 2005) and for the expression of an autosomal homolog of FMR1, FXR2, which encodes a protein whose function is thought to partially overlap with FMRP (Mahishi and Usdin, 2006). [unreadable] [unreadable] The 2nd approach as been to study the molecular events leading to gene silencing on FXS alleles. Previous work has demonstrated that these alleles are heavily methylated. A DNA methyltransferase inhibitor, 5-aza-2-deoxycytidine (5-aza-dC), has been shown to partially reverse the FMR1 silencing. However, not only is this compound toxic, but it requires DNA replication to be effective. It is thus not likely to be useful in post-mitotic cells like neurons where the FMRP deficiency is most apparent. It is also known that the FMR1 promoter in FXS cells is associated with histones characteristic of silenced genes including elevated levels of histone H3 dimethylated at lysine 9 and hypomethylated at lysine 4. Using chromatin immunoprecipitation we have now identified another histone modification associated with silenced alleles. We have also identified a class of small molecules that is able to block this modification. Doing so significantly reverses the gene silencing in patient cells. The mechanism of action of this group of molecules suggests that, unlike 5-aza-dC, they have the potential to be effective in neurons.[unreadable] [unreadable] We have identified the enzyme responsible for the new modification as well as the enzyme involved in the reactivation of the gene. We have also shown that 5-aza-dC probably exerts its effect, either directly or indirectly, via the same histone modification. [unreadable] [unreadable] Transcripts with long CGG-tracts are translated only poorly in some cells, probably because of hairpin formation by the CGG-repeats (Handa, Saha and Usdin, 2003). This has led to the suggestion that gene reactivation alone may not be sufficient to neutralize the effect of the repeats. However, we have shown that the negative effect of these repeats on translation shows tremendous regional differences in the brain (Entezam et. al., 2007). In addition, rare individuals with unsilenced large repeats can have IQs that fall into the normal range. This suggests that translation of transcriptionally active FXS alleles is adequate in those areas of the brain important for learning and memory. These results suggest that it may some day be possible to use the class of compounds we have identified to alleviate some of the more serious symptoms of FXS.